初始化项目，由ModelHub XC社区提供模型
Model: golaxy/gogpt2-13b Source: Original Platform
This commit is contained in:
ModelHub XC
35
.gitattributes
vendored
Normal file
35
.gitattributes
vendored
Normal file
@@ -0,0 +1,35 @@
|
||||
*.7z filter=lfs diff=lfs merge=lfs -text
|
||||
*.arrow filter=lfs diff=lfs merge=lfs -text
|
||||
*.bin filter=lfs diff=lfs merge=lfs -text
|
||||
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
||||
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
||||
*.ftz filter=lfs diff=lfs merge=lfs -text
|
||||
*.gz filter=lfs diff=lfs merge=lfs -text
|
||||
*.h5 filter=lfs diff=lfs merge=lfs -text
|
||||
*.joblib filter=lfs diff=lfs merge=lfs -text
|
||||
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
||||
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
||||
*.model filter=lfs diff=lfs merge=lfs -text
|
||||
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
||||
*.npy filter=lfs diff=lfs merge=lfs -text
|
||||
*.npz filter=lfs diff=lfs merge=lfs -text
|
||||
*.onnx filter=lfs diff=lfs merge=lfs -text
|
||||
*.ot filter=lfs diff=lfs merge=lfs -text
|
||||
*.parquet filter=lfs diff=lfs merge=lfs -text
|
||||
*.pb filter=lfs diff=lfs merge=lfs -text
|
||||
*.pickle filter=lfs diff=lfs merge=lfs -text
|
||||
*.pkl filter=lfs diff=lfs merge=lfs -text
|
||||
*.pt filter=lfs diff=lfs merge=lfs -text
|
||||
*.pth filter=lfs diff=lfs merge=lfs -text
|
||||
*.rar filter=lfs diff=lfs merge=lfs -text
|
||||
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
||||
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
||||
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
||||
*.tar filter=lfs diff=lfs merge=lfs -text
|
||||
*.tflite filter=lfs diff=lfs merge=lfs -text
|
||||
*.tgz filter=lfs diff=lfs merge=lfs -text
|
||||
*.wasm filter=lfs diff=lfs merge=lfs -text
|
||||
*.xz filter=lfs diff=lfs merge=lfs -text
|
||||
*.zip filter=lfs diff=lfs merge=lfs -text
|
||||
*.zst filter=lfs diff=lfs merge=lfs -text
|
||||
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
||||
191
README.md
Normal file
191
README.md
Normal file
@@ -0,0 +1,191 @@
|
||||
---
|
||||
license: apache-2.0
|
||||
datasets:
|
||||
- c-s-ale/alpaca-gpt4-data-zh
|
||||
language:
|
||||
- zh
|
||||
tags:
|
||||
- llama2
|
||||
- chinese-llama2
|
||||
- gogpt2-7b
|
||||
---
|
||||
|
||||
|
||||
# GoGPT2-13B: 基于Llama2-13b训练的中英文增强大模型
|
||||
|
||||
|
||||
|
||||
|
||||
<p align="center">
|
||||
<img alt="GitHub" src="https://img.shields.io/github/license/ymcui/Chinese-LLaMA-Alpaca.svg?color=blue&style=flat-square">
|
||||
<img alt="GitHub top language" src="https://img.shields.io/github/languages/top/ymcui/Chinese-LLaMA-Alpaca">
|
||||
</p>
|
||||
|
||||
> ICT中英文底座增强大模型：70亿参数、130亿参数
|
||||
|
||||
🤗Huggingface上提供了GoGPT权重，目前开放了gogpt-7b和gogpt2-7b权重
|
||||
|
||||
| 模型名称 | 基座模型 | 模型大小 | 下载地址 |
|
||||
|-------------------------------------------------------------|-----------|------|-------------------------------------------------|
|
||||
| [golaxy/gogpt-7b](https://huggingface.co/golaxy/gogpt-7b) | Llama-7b | 7B | [模型下载](https://huggingface.co/golaxy/gogpt-7b) |
|
||||
| [golaxy/gogpt2-7b](https://huggingface.co/golaxy/gogpt2-7b) | Llama2-7b | 7B | [模型下载](https://huggingface.co/golaxy/gogpt2-7b) |
|
||||
| [golaxy/gogpt2-7b-pretrain](https://huggingface.co/golaxy/gogpt2-7b-pretrain) | Llama2-7b | 7B | [模型下载](https://huggingface.co/golaxy/gogpt2-7b-pretrain) |
|
||||
| [golaxy/gogpt2-13b-pretrain](https://huggingface.co/golaxy/gogpt2-13b-pretrain) | Llama2-13b | 13B | [模型下载](https://huggingface.co/golaxy/gogpt2-13b-pretrain) |
|
||||
| [golaxy/gogpt2-13b](https://huggingface.co/golaxy/gogpt2-13b) | Llama2-13b | 13B | [模型下载](https://huggingface.co/golaxy/gogpt2-13b) |
|
||||
|
||||
[GoGPT-Github](https://github.com/yanqiangmiffy/GoGPT)
|
||||
|
||||
|
||||
|
||||
## 测试效果
|
||||
|
||||
- 怎么学习机器学习
|
||||
|
||||
```
|
||||
好的，以下是一些学习机器学习的方法：
|
||||
|
||||
1. 上下线：通过实践项目来学习，例如使用开源机器学习框架（如TensorFlow、Keras、Scikit-Learn等）来实现各种机器学习任务。
|
||||
|
||||
2. 课程学习：选择一些在线或线下的机器学习课程，例如Coursera、Udemy、edX等，这些平台都提供了各种机器学习课程。
|
||||
|
||||
3. 阅读书籍：选择一些经典的机器学习书籍，例如Andrew Ng的《深度学习》、《机器学习》等，这些书籍可以帮助你深入地学习机器学习基础知识。
|
||||
|
||||
4. 加入社区：加入机器学习社区（如Kaggle、Github、Reddit等），与其他机器学习爱好者交流和分享经验，从而不断学习和改进。
|
||||
|
||||
5. 参加比赛：参加一些机器学习比赛（如Kaggle等），可以通过解决实际问题来学习并提高自己的技能。
|
||||
|
||||
总之，学习机器学习需要不断实践和探索，找到适合自己的学习方法。
|
||||
```
|
||||
|
||||
- 撰写一篇交响乐音乐会评论，讨论乐团的表演和观众的整体体验。
|
||||
|
||||
```
|
||||
交响乐音乐会是音乐爱好者和古典音乐的爱好者的梦想。上周末，我参加了一场交响乐音乐会，我的经历真的很难忘。
|
||||
|
||||
首先，乐团的表演非常出色。他们都是专业音乐家，他们的演奏技巧非常娴熟。他们在演奏时，展现出了音乐的真谛，让人感受到音乐的力量和魅力。同时，他们还非常出色地演绎了交响乐中的不同元素，如高音、低音、节奏、调性和变化等。他们的表演非常自然，没有过分的紧张或放松，给人一种平和的、和谐的感觉。
|
||||
|
||||
其次，观众的整体体验也非常不错。音乐会现场非常宽敞明亮，座椅间隔非常宽敞。观众可以自由地欣赏音乐的每一个细节，同时还可以享受到音乐的魅力。同时，音乐会现场的氛围也非常友好，观众之间相互交流，享受音乐的共品。此外，音乐会现场还设有各种美食和饮料供观众品尝，增加了观众的舒适度。
|
||||
|
||||
总的来说，这次交响乐音乐会是一次非常愉快的音乐体验。乐团的表演非常出色，观众的整体体验也非常不错。我真的非常期待下一场音乐会，再次领略到音乐的魅力。
|
||||
```
|
||||
|
||||
- 为一部最近上映的科幻电影撰写一个引人入胜的影评，讨论它的情节，角色和特效。
|
||||
```
|
||||
《星球大战：新纪元》是一部令人兴奋的科幻电影，它延续了之前的星球大战系列的故事情节和角色，同时也带来了新的挑战和惊喜。
|
||||
|
||||
这部电影的情节非常引人入胜，它讲述了一群反抗军与帝国之间的斗争，反抗军试图推翻帝国的统治，而帝国则竭力维持其统治。故事情节的复杂程度和紧张程度让人惊叹，每个角色都有自己独特的动机和经历，这使得整个电影的故事更加生动。
|
||||
|
||||
这部电影的主要角色包括雷伊（Rey），一个年轻的女武士，以及她的导师凯诺（Kylo Ren），一名邪恶的武士。这两个角色的互动和冲突是电影中的一个重要焦点，他们的故事情节充满了神秘和悬念，使得观众期待着下一刻会发生什么。此外，电影中还有一些新角色的出现，如米斯·恩德（Maisy Endor），一个新的星球上反抗军的领导者，以及C-3PO和R2-D2等经典的机器人角色，这些角色的回归和表现也都令人兴奋。
|
||||
|
||||
这部电影的视觉效果也非常惊人，它使用了最新的技术和特效来营造出逼真的宇宙场景和战斗场面。电影中的星球、建筑物、机械设备和武器的视觉效果都非常逼真，而且动作场面也非常刺激。电影中的视觉效果让人感到惊奇，也使观众更加沉浸在电影中的世界。
|
||||
|
||||
总的来说，《星球大战：新纪元》是一部非常值得一看的科幻电影，它讲述了一个引人入胜的故事，有着精彩的角色和特效，使得整个电影令人兴奋和震撼。这部电影的成功，也预示着星球大战系列的未来。
|
||||
```
|
||||
|
||||
|
||||
- 人类平均一生眨眼多少次？请尝试解释你的答案。你的解释应该逐步带领读者理解你的推理过程
|
||||
```
|
||||
人类平均一生眨眼几次是5600000次左右。这个数字是由许多不同的研究结果和估算得出的。
|
||||
|
||||
首先，眨眼频率随年龄的增加而变化。婴儿眨眼频率比成人高，随着年龄的增长，眨眼频率逐渐下降。根据这个规律，我们可以计算出成人眨眼频率约为10次/分钟。
|
||||
|
||||
其次，眨眼频率还受到多种因素的影响。例如，当人们感到紧张或疲劳时，眨眼频率会增加。同样地，当人们感到愉悦或放松时，眨眼频率会降低。此外，不同的身体部位的眨眼频率也不同。例如，眼睛内侧的眨眼频率比内侧的眨眼频率高。
|
||||
|
||||
最后，由于人们的眼睛健康状况也会影响眨眼频率，例如眼底病变或视力问题。这些因素都可以导致眨眼频率的改变。
|
||||
|
||||
综上所述，人类平均一生眨眼几次约为5600000次左右。这个数字是由多个因素影响的结果，包括年龄、身体状况、情感状态等。
|
||||
```
|
||||
|
||||
- 请用python编程语言实现快速排序算法的代码
|
||||
|
||||
以下是使用Python快速排序算法的示例代码：
|
||||
|
||||
```python
|
||||
def quick_sort(arr):
|
||||
if len(arr) <= 1:
|
||||
return arr
|
||||
else:
|
||||
pivot = arr[0]
|
||||
left = []
|
||||
right = []
|
||||
for i in range(1, len(arr)):
|
||||
if arr[i] < pivot:
|
||||
left.append(arr[i])
|
||||
else:
|
||||
right.append(arr[i])
|
||||
return quick_sort(left) + [pivot] + quick_sort(right)
|
||||
```
|
||||
|
||||
这个函数接受一个整数数组作为输入，并在最坏情况下的时间复杂度为O(n^2)的情况下，最佳情况下时间复杂度为O(nlogn)。
|
||||
|
||||
|
||||
- 如何提高我的时间管理技能？
|
||||
```
|
||||
以下是一些提高时间管理技能的建议：
|
||||
|
||||
1. 制定计划：每天早上或每天开始时，制定一个详细的计划，包括每天需要完成的任务和目标。
|
||||
|
||||
2. 优先级排序：将任务按照优先级排序，先完成重要任务。
|
||||
|
||||
3. 集中注意力：集中注意力，将注意力集中在当前的任务上，避免分心。
|
||||
|
||||
4. 避免拖延：尽可能避免拖延，尽快完成任务，避免任务过期。
|
||||
|
||||
5. 避免多任务：多任务会增加任务的时间，所以最好只专注一个任务，直到完成它。
|
||||
|
||||
6. 休息：休息是恢复和提高生产力的关键，因此定期休息，放松身心。
|
||||
|
||||
7. 调整时间：如果发现自己的时间管理有问题，可以调整时间，例如延长休息时间或缩短任务时间。
|
||||
|
||||
8. 学习新技能：学习新的技能可以帮助提高工作效率，例如使用工具或软件来管理任务。
|
||||
|
||||
9. 寻求帮助：如果你发现自己难以管理时间，可以寻求家人、朋友或同事的帮助。
|
||||
|
||||
希望这些建议可以帮助您提高时间管理技能。
|
||||
```
|
||||
|
||||
- Python和JavaScript编程语言的主要区别是什么？
|
||||
```
|
||||
Python和JavaScript是两个不同的编程语言，它们有以下主要区别：
|
||||
|
||||
1. 数据类型：Python是一种动态类型的编程语言，它无法指定数据类型，而是根据变量中使用的值自动确定数据类型。JavaScript是一种动态类型编程语言，它使用字面量来指定变量的数据类型。
|
||||
|
||||
2. 语法：Python和JavaScript的语法有一些相似之处，但也有一些不同之处。例如，Python使用缩进来定义代码块，而JavaScript使用花括号。
|
||||
|
||||
3. 执行环境：Python通常运行在本地计算机的Python解释器中运行，而JavaScript通常运行在浏览器中运行。这意味着Python通常用于开发桌面应用程序和Web服务器端应用程序，而JavaScript主要用于开发Web应用程序。
|
||||
|
||||
4. 编程语言特性：Python具有更高的可读性和可维护性，而JavaScript具有更好的可扩展性和可维护性。
|
||||
|
||||
总之，Python和JavaScript是两种不同的编程语言，具有不同的特点和应用场景。选择哪种编程语言取决于项目的具体需求和开发者的技能水平。
|
||||
```
|
||||
- 植物蛋白和动物蛋白的主要区别是什么？
|
||||
```
|
||||
植物蛋白和动物蛋白的主要区别在于来源和氨基酸成分。
|
||||
|
||||
来源：
|
||||
植物蛋白主要来自植物组织，如豆类、种子、果实、叶子等。而动物蛋白主要来自动物的身体部位，如肌肉、内脏、鱼类、蛋类等。
|
||||
|
||||
氨基酸成分：
|
||||
植物蛋白中的主要氨基酸是植物性氨基酸，如氨基酸，植物性必需氨基酸（如硫胺素和植酸）等。而动物蛋白中的主要氨基酸是动物性氨基酸，如肌肉蛋白、乳糖蛋白等。
|
||||
|
||||
因此，植物蛋白和动物蛋白的营养价值是不同的。植物蛋白富含植物性必需氨基酸，如硫胺素、植酸等，而动物蛋白则含有更多的动物性必需氨基酸，如肌肉蛋白、乳糖蛋白等。植物蛋白通常比动物蛋白低热量，但也有些植物蛋白的蛋白质质量较低。动物蛋白通常比植物蛋白高蛋白质质量，但同样含有较多的饱和脂肪和胆固醇等有害物质。
|
||||
```
|
||||
|
||||
- 影响消费者行为的主要因素是什么？
|
||||
```
|
||||
影响消费者行为的因素有很多，以下是其中几个主要的因素：
|
||||
|
||||
1. 消费需求：消费者的需求是影响消费者行为的主要因素之一。消费者的需求是根据个人利益和价值观而产生的。例如，如果一个消费者喜欢时尚的服装，那么他们可能会对购买这些服装的商场和品牌的选择更有吸引力。
|
||||
|
||||
2. 购买力：消费者购买力的限制是一个影响消费者行为的因素。消费者可能会因为预算限制而选择购买某些商品。
|
||||
|
||||
3. 价格：价格也是影响消费者行为的因素之一。消费者通常会比较不同商品的价格，并根据价格选择购买哪种商品。
|
||||
|
||||
4. 品牌知名度：品牌知名度是影响消费者行为的因素之一。消费者通常会选择有信誉的品牌，因为他们相信这些品牌能够提供高质量的产品和服务。
|
||||
|
||||
5. 社交媒体：社交媒体在影响消费者行为中扮演越来越重要的作用。许多消费者会在社交媒体上了解商品和品牌的信息，并选择购买这些商品。
|
||||
|
||||
6. 社交媒体：社交媒体在影响消费者行为中扮演越来越重要的角色。许多消费者会在社交媒体上了解商品和品牌的信息，并选择购买这些商品。
|
||||
|
||||
总之，影响消费者行为的主要因素有很多，这些因素往往是相互关联的。企业可以通过了解这些因素，来制定有效的营销策略，吸引和留住消费者。
|
||||
```
|
||||
3
added_tokens.json
Normal file
3
added_tokens.json
Normal file
@@ -0,0 +1,3 @@
|
||||
{
|
||||
"[PAD]": 68419
|
||||
}
|
||||
26
config.json
Normal file
26
config.json
Normal file
@@ -0,0 +1,26 @@
|
||||
{
|
||||
"_name_or_path": "/data/searchgpt/yq/Firefly/llama2-13b-moss-sft",
|
||||
"architectures": [
|
||||
"LlamaForCausalLM"
|
||||
],
|
||||
"bos_token_id": 1,
|
||||
"eos_token_id": 2,
|
||||
"hidden_act": "silu",
|
||||
"hidden_size": 5120,
|
||||
"initializer_range": 0.02,
|
||||
"intermediate_size": 13824,
|
||||
"max_position_embeddings": 4096,
|
||||
"model_type": "llama",
|
||||
"num_attention_heads": 40,
|
||||
"num_hidden_layers": 40,
|
||||
"num_key_value_heads": 40,
|
||||
"pad_token_id": 0,
|
||||
"pretraining_tp": 1,
|
||||
"rms_norm_eps": 1e-05,
|
||||
"rope_scaling": null,
|
||||
"tie_word_embeddings": false,
|
||||
"torch_dtype": "float16",
|
||||
"transformers_version": "4.31.0",
|
||||
"use_cache": false,
|
||||
"vocab_size": 68420
|
||||
}
|
||||
8
generation_config.json
Normal file
8
generation_config.json
Normal file
@@ -0,0 +1,8 @@
|
||||
{
|
||||
"_from_model_config": true,
|
||||
"bos_token_id": 1,
|
||||
"eos_token_id": 2,
|
||||
"pad_token_id": 0,
|
||||
"transformers_version": "4.31.0",
|
||||
"use_cache": false
|
||||
}
|
||||
3
pytorch_model-00001-of-00006.bin
Normal file
3
pytorch_model-00001-of-00006.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:e20bb11669998a4eadd2e7e4e0386c4b4357c436ea8781d3fbb40a43a8fdb013
|
||||
size 9999835619
|
||||
3
pytorch_model-00002-of-00006.bin
Normal file
3
pytorch_model-00002-of-00006.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:6ff29ffbe0ba2a3f3a3aa3553b50419e2e7a87dd2f1a27c63b177b0223cc8f5d
|
||||
size 9867456961
|
||||
3
pytorch_model-00003-of-00006.bin
Normal file
3
pytorch_model-00003-of-00006.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:fa4bfcb0b93b4dcad872eba1c1379c1fce05bf066b86181667c3df200572aa6b
|
||||
size 9867456973
|
||||
3
pytorch_model-00004-of-00006.bin
Normal file
3
pytorch_model-00004-of-00006.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:3a03304b49372511858d8299ba437f5287201f331de4aa28d8f54536fbd92e17
|
||||
size 9940856385
|
||||
3
pytorch_model-00005-of-00006.bin
Normal file
3
pytorch_model-00005-of-00006.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:e36497143e986b06ea70c0629b0b3580993a89c9002e5b8eeee97cfc50591e39
|
||||
size 9940856943
|
||||
3
pytorch_model-00006-of-00006.bin
Normal file
3
pytorch_model-00006-of-00006.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:3cd0b4b5bd6b894937ad75b20cc3158e6e0f40229e3b81b8896e07b67a91865f
|
||||
size 3938905883
|
||||
410
pytorch_model.bin.index.json
Normal file
410
pytorch_model.bin.index.json
Normal file
@@ -0,0 +1,410 @@
|
||||
{
|
||||
"metadata": {
|
||||
"total_size": 53555230720
|
||||
},
|
||||
"weight_map": {
|
||||
"lm_head.weight": "pytorch_model-00006-of-00006.bin",
|
||||
"model.embed_tokens.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.0.input_layernorm.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.0.mlp.down_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.0.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.0.mlp.up_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.0.post_attention_layernorm.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.0.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.0.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.0.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.0.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.0.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.1.input_layernorm.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.1.mlp.down_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.1.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.1.mlp.up_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.1.post_attention_layernorm.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.1.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.1.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.1.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.1.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.1.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.10.input_layernorm.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.10.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.10.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.10.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.10.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.10.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.10.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.10.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.10.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.10.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.11.input_layernorm.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.11.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.11.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.11.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.11.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.11.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.11.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.11.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.11.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.11.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.12.input_layernorm.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.12.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.12.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.12.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.12.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.12.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.12.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.12.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.12.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.12.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.13.input_layernorm.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.13.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.13.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.13.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.13.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.13.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.13.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.13.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.13.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.13.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.14.input_layernorm.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.14.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.14.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.14.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.14.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.14.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.14.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.14.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.14.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.14.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.15.input_layernorm.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.15.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.15.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.15.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.15.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.15.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.15.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.15.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.15.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.15.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.16.input_layernorm.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.16.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.16.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.16.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.16.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.16.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.16.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.16.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.16.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.16.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.17.input_layernorm.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.17.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.17.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.17.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.17.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.17.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.17.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.17.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.17.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.17.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.18.input_layernorm.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.18.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.18.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.18.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.18.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.18.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.18.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.18.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.18.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.18.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.19.input_layernorm.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.19.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.19.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.19.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.19.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.19.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.19.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.19.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.19.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.19.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.2.input_layernorm.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.2.mlp.down_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.2.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.2.mlp.up_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.2.post_attention_layernorm.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.2.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.2.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.2.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.2.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.2.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.20.input_layernorm.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.20.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.20.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.20.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.20.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.20.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.20.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.20.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.20.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.20.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.21.input_layernorm.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.21.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.21.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.21.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.21.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.21.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.21.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.21.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.21.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.21.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.22.input_layernorm.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.22.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.22.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.22.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.22.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.22.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.22.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.22.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.22.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.22.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
|
||||
"model.layers.23.input_layernorm.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.23.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.23.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.23.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.23.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.23.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.23.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.23.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.23.self_attn.rotary_emb.inv_freq": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.23.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.24.input_layernorm.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.24.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.24.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.24.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.24.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.24.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.24.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.24.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.24.self_attn.rotary_emb.inv_freq": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.24.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.25.input_layernorm.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.25.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.25.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.25.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.25.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.25.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.25.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.25.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.25.self_attn.rotary_emb.inv_freq": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.25.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.26.input_layernorm.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.26.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.26.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.26.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.26.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.26.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.26.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.26.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.26.self_attn.rotary_emb.inv_freq": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.26.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.27.input_layernorm.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.27.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.27.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.27.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.27.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.27.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.27.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.27.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.27.self_attn.rotary_emb.inv_freq": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.27.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.28.input_layernorm.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.28.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.28.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.28.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.28.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.28.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.28.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.28.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.28.self_attn.rotary_emb.inv_freq": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.28.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.29.input_layernorm.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.29.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.29.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.29.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.29.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.29.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.29.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.29.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.29.self_attn.rotary_emb.inv_freq": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.29.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.3.input_layernorm.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.3.mlp.down_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.3.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.3.mlp.up_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.3.post_attention_layernorm.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.3.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.3.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.3.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.3.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.3.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.30.input_layernorm.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.30.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.30.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.30.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.30.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.30.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.30.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.30.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
|
||||
"model.layers.30.self_attn.rotary_emb.inv_freq": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.30.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.31.input_layernorm.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.31.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.31.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.31.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.31.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.31.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.31.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.31.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.31.self_attn.rotary_emb.inv_freq": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.31.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.32.input_layernorm.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.32.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.32.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.32.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.32.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.32.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.32.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.32.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.32.self_attn.rotary_emb.inv_freq": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.32.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.33.input_layernorm.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.33.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.33.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.33.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.33.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.33.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.33.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.33.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.33.self_attn.rotary_emb.inv_freq": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.33.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.34.input_layernorm.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.34.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.34.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.34.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.34.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.34.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.34.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.34.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.34.self_attn.rotary_emb.inv_freq": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.34.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.35.input_layernorm.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.35.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.35.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.35.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.35.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.35.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.35.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.35.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.35.self_attn.rotary_emb.inv_freq": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.35.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.36.input_layernorm.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.36.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.36.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.36.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.36.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.36.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.36.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.36.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.36.self_attn.rotary_emb.inv_freq": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.36.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.37.input_layernorm.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.37.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.37.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.37.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.37.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.37.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.37.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.37.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.37.self_attn.rotary_emb.inv_freq": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.37.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin",
|
||||
"model.layers.38.input_layernorm.weight": "pytorch_model-00006-of-00006.bin",
|
||||
"model.layers.38.mlp.down_proj.weight": "pytorch_model-00006-of-00006.bin",
|
||||
"model.layers.38.mlp.gate_proj.weight": "pytorch_model-00006-of-00006.bin",
|
||||
"model.layers.38.mlp.up_proj.weight": "pytorch_model-00006-of-00006.bin",
|
||||
"model.layers.38.post_attention_layernorm.weight": "pytorch_model-00006-of-00006.bin",
|
||||
"model.layers.38.self_attn.k_proj.weight": "pytorch_model-00006-of-00006.bin",
|
||||
"model.layers.38.self_attn.o_proj.weight": "pytorch_model-00006-of-00006.bin",
|
||||
"model.layers.38.self_attn.q_proj.weight": "pytorch_model-00006-of-00006.bin",
|
||||
"model.layers.38.self_attn.rotary_emb.inv_freq": "pytorch_model-00006-of-00006.bin",
|
||||
"model.layers.38.self_attn.v_proj.weight": "pytorch_model-00006-of-00006.bin",
|
||||
"model.layers.39.input_layernorm.weight": "pytorch_model-00006-of-00006.bin",
|
||||
"model.layers.39.mlp.down_proj.weight": "pytorch_model-00006-of-00006.bin",
|
||||
"model.layers.39.mlp.gate_proj.weight": "pytorch_model-00006-of-00006.bin",
|
||||
"model.layers.39.mlp.up_proj.weight": "pytorch_model-00006-of-00006.bin",
|
||||
"model.layers.39.post_attention_layernorm.weight": "pytorch_model-00006-of-00006.bin",
|
||||
"model.layers.39.self_attn.k_proj.weight": "pytorch_model-00006-of-00006.bin",
|
||||
"model.layers.39.self_attn.o_proj.weight": "pytorch_model-00006-of-00006.bin",
|
||||
"model.layers.39.self_attn.q_proj.weight": "pytorch_model-00006-of-00006.bin",
|
||||
"model.layers.39.self_attn.rotary_emb.inv_freq": "pytorch_model-00006-of-00006.bin",
|
||||
"model.layers.39.self_attn.v_proj.weight": "pytorch_model-00006-of-00006.bin",
|
||||
"model.layers.4.input_layernorm.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.4.mlp.down_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.4.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.4.mlp.up_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.4.post_attention_layernorm.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.4.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.4.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.4.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.4.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.4.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.5.input_layernorm.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.5.mlp.down_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.5.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.5.mlp.up_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.5.post_attention_layernorm.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.5.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.5.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.5.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.5.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.5.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.6.input_layernorm.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.6.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.6.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.6.mlp.up_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.6.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.6.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.6.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.6.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.6.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.6.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin",
|
||||
"model.layers.7.input_layernorm.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.7.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.7.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.7.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.7.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.7.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.7.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.7.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.7.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.7.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.8.input_layernorm.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.8.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.8.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.8.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.8.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.8.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.8.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.8.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.8.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.8.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.9.input_layernorm.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.9.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.9.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.9.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.9.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.9.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.9.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.9.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.9.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00006.bin",
|
||||
"model.layers.9.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
|
||||
"model.norm.weight": "pytorch_model-00006-of-00006.bin"
|
||||
}
|
||||
}
|
||||
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:ce7d4637b767412387e2b81e164bfafa403b7f58202f069ab6c4f05967ace0df
|
||||
size 1451004
|
||||
24
special_tokens_map.json
Normal file
24
special_tokens_map.json
Normal file
@@ -0,0 +1,24 @@
|
||||
{
|
||||
"bos_token": {
|
||||
"content": "<s>",
|
||||
"lstrip": false,
|
||||
"normalized": true,
|
||||
"rstrip": false,
|
||||
"single_word": false
|
||||
},
|
||||
"eos_token": {
|
||||
"content": "</s>",
|
||||
"lstrip": false,
|
||||
"normalized": true,
|
||||
"rstrip": false,
|
||||
"single_word": false
|
||||
},
|
||||
"pad_token": "[PAD]",
|
||||
"unk_token": {
|
||||
"content": "<unk>",
|
||||
"lstrip": false,
|
||||
"normalized": true,
|
||||
"rstrip": false,
|
||||
"single_word": false
|
||||
}
|
||||
}
|
||||
165805
tokenizer.json
Normal file
165805
tokenizer.json
Normal file
File diff suppressed because it is too large
Load Diff
3
tokenizer.model
Normal file
3
tokenizer.model
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:bd5ab2c18ed07a14f3aa55518dcf08bbee4fe86c9423e86ba61f60a82ab31fa7
|
||||
size 1077901
|
||||
37
tokenizer_config.json
Normal file
37
tokenizer_config.json
Normal file
@@ -0,0 +1,37 @@
|
||||
{
|
||||
"add_bos_token": true,
|
||||
"add_eos_token": false,
|
||||
"bos_token": {
|
||||
"__type": "AddedToken",
|
||||
"content": "<s>",
|
||||
"lstrip": false,
|
||||
"normalized": true,
|
||||
"rstrip": false,
|
||||
"single_word": false
|
||||
},
|
||||
"clean_up_tokenization_spaces": false,
|
||||
"eos_token": {
|
||||
"__type": "AddedToken",
|
||||
"content": "</s>",
|
||||
"lstrip": false,
|
||||
"normalized": true,
|
||||
"rstrip": false,
|
||||
"single_word": false
|
||||
},
|
||||
"legacy": true,
|
||||
"model_max_length": 4096,
|
||||
"pad_token": null,
|
||||
"padding_side": "right",
|
||||
"sp_model_kwargs": {},
|
||||
"tokenizer_class": "LlamaTokenizer",
|
||||
"trust_remote_code": true,
|
||||
"unk_token": {
|
||||
"__type": "AddedToken",
|
||||
"content": "<unk>",
|
||||
"lstrip": false,
|
||||
"normalized": true,
|
||||
"rstrip": false,
|
||||
"single_word": false
|
||||
},
|
||||
"use_fast": false
|
||||
}
|
||||
54016
trainer_state.json
Normal file
54016
trainer_state.json
Normal file
File diff suppressed because it is too large
Load Diff
3
training_args.bin
Normal file
3
training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:57926aa6785cbb15971c4029ef163a66d5c710e5a102c59a50c20779bd8c5da6
|
||||
size 5947
|
||||
578
zero_to_fp32.py
Executable file
578
zero_to_fp32.py
Executable file
@@ -0,0 +1,578 @@
|
||||
#!/usr/bin/env python
|
||||
|
||||
# Copyright (c) Microsoft Corporation.
|
||||
# SPDX-License-Identifier: Apache-2.0
|
||||
|
||||
# DeepSpeed Team
|
||||
|
||||
# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
|
||||
# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
|
||||
# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
|
||||
# application.
|
||||
#
|
||||
# example: python zero_to_fp32.py . pytorch_model.bin
|
||||
|
||||
import argparse
|
||||
import torch
|
||||
import glob
|
||||
import math
|
||||
import os
|
||||
import re
|
||||
from collections import OrderedDict
|
||||
from dataclasses import dataclass
|
||||
|
||||
# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
|
||||
# DeepSpeed data structures it has to be available in the current python environment.
|
||||
from deepspeed.utils import logger
|
||||
from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
|
||||
FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
|
||||
FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
|
||||
|
||||
|
||||
@dataclass
|
||||
class zero_model_state:
|
||||
buffers: dict()
|
||||
param_shapes: dict()
|
||||
shared_params: list
|
||||
ds_version: int
|
||||
frozen_param_shapes: dict()
|
||||
frozen_param_fragments: dict()
|
||||
|
||||
|
||||
debug = 0
|
||||
|
||||
# load to cpu
|
||||
device = torch.device('cpu')
|
||||
|
||||
|
||||
def atoi(text):
|
||||
return int(text) if text.isdigit() else text
|
||||
|
||||
|
||||
def natural_keys(text):
|
||||
'''
|
||||
alist.sort(key=natural_keys) sorts in human order
|
||||
http://nedbatchelder.com/blog/200712/human_sorting.html
|
||||
(See Toothy's implementation in the comments)
|
||||
'''
|
||||
return [atoi(c) for c in re.split(r'(\d+)', text)]
|
||||
|
||||
|
||||
def get_model_state_file(checkpoint_dir, zero_stage):
|
||||
if not os.path.isdir(checkpoint_dir):
|
||||
raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
|
||||
|
||||
# there should be only one file
|
||||
if zero_stage <= 2:
|
||||
file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
|
||||
elif zero_stage == 3:
|
||||
file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
|
||||
|
||||
if not os.path.exists(file):
|
||||
raise FileNotFoundError(f"can't find model states file at '{file}'")
|
||||
|
||||
return file
|
||||
|
||||
|
||||
def get_checkpoint_files(checkpoint_dir, glob_pattern):
|
||||
# XXX: need to test that this simple glob rule works for multi-node setup too
|
||||
ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
|
||||
|
||||
if len(ckpt_files) == 0:
|
||||
raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
|
||||
|
||||
return ckpt_files
|
||||
|
||||
|
||||
def get_optim_files(checkpoint_dir):
|
||||
return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
|
||||
|
||||
|
||||
def get_model_state_files(checkpoint_dir):
|
||||
return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
|
||||
|
||||
|
||||
def parse_model_states(files):
|
||||
zero_model_states = []
|
||||
for file in files:
|
||||
state_dict = torch.load(file, map_location=device)
|
||||
|
||||
if BUFFER_NAMES not in state_dict:
|
||||
raise ValueError(f"{file} is not a model state checkpoint")
|
||||
buffer_names = state_dict[BUFFER_NAMES]
|
||||
if debug:
|
||||
print("Found buffers:", buffer_names)
|
||||
|
||||
# recover just the buffers while restoring them to fp32 if they were saved in fp16
|
||||
buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
|
||||
param_shapes = state_dict[PARAM_SHAPES]
|
||||
|
||||
# collect parameters that are included in param_shapes
|
||||
param_names = []
|
||||
for s in param_shapes:
|
||||
for name in s.keys():
|
||||
param_names.append(name)
|
||||
|
||||
# update with frozen parameters
|
||||
frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
|
||||
if frozen_param_shapes is not None:
|
||||
if debug:
|
||||
print(f"Found frozen_param_shapes: {frozen_param_shapes}")
|
||||
param_names += list(frozen_param_shapes.keys())
|
||||
|
||||
# handle shared params
|
||||
shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
|
||||
|
||||
ds_version = state_dict.get(DS_VERSION, None)
|
||||
|
||||
frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
|
||||
|
||||
z_model_state = zero_model_state(buffers=buffers,
|
||||
param_shapes=param_shapes,
|
||||
shared_params=shared_params,
|
||||
ds_version=ds_version,
|
||||
frozen_param_shapes=frozen_param_shapes,
|
||||
frozen_param_fragments=frozen_param_fragments)
|
||||
zero_model_states.append(z_model_state)
|
||||
|
||||
return zero_model_states
|
||||
|
||||
|
||||
def parse_optim_states(files, ds_checkpoint_dir):
|
||||
|
||||
total_files = len(files)
|
||||
state_dicts = []
|
||||
for f in files:
|
||||
state_dicts.append(torch.load(f, map_location=device))
|
||||
|
||||
if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
|
||||
raise ValueError(f"{files[0]} is not a zero checkpoint")
|
||||
zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
|
||||
world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
|
||||
|
||||
# For ZeRO-2 each param group can have different partition_count as data parallelism for expert
|
||||
# parameters can be different from data parallelism for non-expert parameters. So we can just
|
||||
# use the max of the partition_count to get the dp world_size.
|
||||
|
||||
if type(world_size) is list:
|
||||
world_size = max(world_size)
|
||||
|
||||
if world_size != total_files:
|
||||
raise ValueError(
|
||||
f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
|
||||
"Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
|
||||
)
|
||||
|
||||
# the groups are named differently in each stage
|
||||
if zero_stage <= 2:
|
||||
fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
|
||||
elif zero_stage == 3:
|
||||
fp32_groups_key = FP32_FLAT_GROUPS
|
||||
else:
|
||||
raise ValueError(f"unknown zero stage {zero_stage}")
|
||||
|
||||
if zero_stage <= 2:
|
||||
fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
|
||||
elif zero_stage == 3:
|
||||
# if there is more than one param group, there will be multiple flattened tensors - one
|
||||
# flattened tensor per group - for simplicity merge them into a single tensor
|
||||
#
|
||||
# XXX: could make the script more memory efficient for when there are multiple groups - it
|
||||
# will require matching the sub-lists of param_shapes for each param group flattened tensor
|
||||
|
||||
fp32_flat_groups = [
|
||||
torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts))
|
||||
]
|
||||
|
||||
return zero_stage, world_size, fp32_flat_groups
|
||||
|
||||
|
||||
def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir):
|
||||
"""
|
||||
Returns fp32 state_dict reconstructed from ds checkpoint
|
||||
|
||||
Args:
|
||||
- ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
|
||||
|
||||
"""
|
||||
print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
|
||||
|
||||
optim_files = get_optim_files(ds_checkpoint_dir)
|
||||
zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
|
||||
print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
|
||||
|
||||
model_files = get_model_state_files(ds_checkpoint_dir)
|
||||
|
||||
zero_model_states = parse_model_states(model_files)
|
||||
print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
|
||||
|
||||
if zero_stage <= 2:
|
||||
return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states)
|
||||
elif zero_stage == 3:
|
||||
return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states)
|
||||
|
||||
|
||||
def _zero2_merge_frozen_params(state_dict, zero_model_states):
|
||||
if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
|
||||
return
|
||||
|
||||
frozen_param_shapes = zero_model_states[0].frozen_param_shapes
|
||||
frozen_param_fragments = zero_model_states[0].frozen_param_fragments
|
||||
|
||||
if debug:
|
||||
num_elem = sum(s.numel() for s in frozen_param_shapes.values())
|
||||
print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
|
||||
|
||||
wanted_params = len(frozen_param_shapes)
|
||||
wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
|
||||
avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
|
||||
print(f'Frozen params: Have {avail_numel} numels to process.')
|
||||
print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
|
||||
|
||||
total_params = 0
|
||||
total_numel = 0
|
||||
for name, shape in frozen_param_shapes.items():
|
||||
total_params += 1
|
||||
unpartitioned_numel = shape.numel()
|
||||
total_numel += unpartitioned_numel
|
||||
|
||||
state_dict[name] = frozen_param_fragments[name]
|
||||
|
||||
if debug:
|
||||
print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
|
||||
|
||||
print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
|
||||
|
||||
|
||||
def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
|
||||
param_shapes = zero_model_states[0].param_shapes
|
||||
|
||||
# Reconstruction protocol:
|
||||
#
|
||||
# XXX: document this
|
||||
|
||||
if debug:
|
||||
for i in range(world_size):
|
||||
for j in range(len(fp32_flat_groups[0])):
|
||||
print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
|
||||
|
||||
# XXX: memory usage doubles here (zero2)
|
||||
num_param_groups = len(fp32_flat_groups[0])
|
||||
merged_single_partition_of_fp32_groups = []
|
||||
for i in range(num_param_groups):
|
||||
merged_partitions = [sd[i] for sd in fp32_flat_groups]
|
||||
full_single_fp32_vector = torch.cat(merged_partitions, 0)
|
||||
merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
|
||||
avail_numel = sum(
|
||||
[full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
|
||||
|
||||
if debug:
|
||||
wanted_params = sum([len(shapes) for shapes in param_shapes])
|
||||
wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
|
||||
# not asserting if there is a mismatch due to possible padding
|
||||
print(f"Have {avail_numel} numels to process.")
|
||||
print(f"Need {wanted_numel} numels in {wanted_params} params.")
|
||||
|
||||
# params
|
||||
# XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
|
||||
# out-of-core computing solution
|
||||
total_numel = 0
|
||||
total_params = 0
|
||||
for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
|
||||
offset = 0
|
||||
avail_numel = full_single_fp32_vector.numel()
|
||||
for name, shape in shapes.items():
|
||||
|
||||
unpartitioned_numel = shape.numel()
|
||||
total_numel += unpartitioned_numel
|
||||
total_params += 1
|
||||
|
||||
if debug:
|
||||
print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
|
||||
state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
|
||||
offset += unpartitioned_numel
|
||||
|
||||
# Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
|
||||
# avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
|
||||
# paddings performed in the code it's almost impossible to predict the exact numbers w/o the
|
||||
# live optimizer object, so we are checking that the numbers are within the right range
|
||||
align_to = 2 * world_size
|
||||
|
||||
def zero2_align(x):
|
||||
return align_to * math.ceil(x / align_to)
|
||||
|
||||
if debug:
|
||||
print(f"original offset={offset}, avail_numel={avail_numel}")
|
||||
|
||||
offset = zero2_align(offset)
|
||||
avail_numel = zero2_align(avail_numel)
|
||||
|
||||
if debug:
|
||||
print(f"aligned offset={offset}, avail_numel={avail_numel}")
|
||||
|
||||
# Sanity check
|
||||
if offset != avail_numel:
|
||||
raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
|
||||
|
||||
print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
|
||||
|
||||
|
||||
def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states):
|
||||
state_dict = OrderedDict()
|
||||
|
||||
# buffers
|
||||
buffers = zero_model_states[0].buffers
|
||||
state_dict.update(buffers)
|
||||
if debug:
|
||||
print(f"added {len(buffers)} buffers")
|
||||
|
||||
_zero2_merge_frozen_params(state_dict, zero_model_states)
|
||||
|
||||
_zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
|
||||
|
||||
# recover shared parameters
|
||||
for pair in zero_model_states[0].shared_params:
|
||||
if pair[1] in state_dict:
|
||||
state_dict[pair[0]] = state_dict[pair[1]]
|
||||
|
||||
return state_dict
|
||||
|
||||
|
||||
def zero3_partitioned_param_info(unpartitioned_numel, world_size):
|
||||
remainder = unpartitioned_numel % world_size
|
||||
padding_numel = (world_size - remainder) if remainder else 0
|
||||
partitioned_numel = math.ceil(unpartitioned_numel / world_size)
|
||||
return partitioned_numel, padding_numel
|
||||
|
||||
|
||||
def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
|
||||
if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
|
||||
return
|
||||
|
||||
if debug:
|
||||
for i in range(world_size):
|
||||
num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
|
||||
print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
|
||||
|
||||
frozen_param_shapes = zero_model_states[0].frozen_param_shapes
|
||||
wanted_params = len(frozen_param_shapes)
|
||||
wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
|
||||
avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
|
||||
print(f'Frozen params: Have {avail_numel} numels to process.')
|
||||
print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
|
||||
|
||||
total_params = 0
|
||||
total_numel = 0
|
||||
for name, shape in zero_model_states[0].frozen_param_shapes.items():
|
||||
total_params += 1
|
||||
unpartitioned_numel = shape.numel()
|
||||
total_numel += unpartitioned_numel
|
||||
|
||||
param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
|
||||
state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
|
||||
|
||||
partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
|
||||
|
||||
if debug:
|
||||
print(
|
||||
f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
|
||||
)
|
||||
|
||||
print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
|
||||
|
||||
|
||||
def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
|
||||
param_shapes = zero_model_states[0].param_shapes
|
||||
avail_numel = fp32_flat_groups[0].numel() * world_size
|
||||
# Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
|
||||
# param, re-consolidating each param, while dealing with padding if any
|
||||
|
||||
# merge list of dicts, preserving order
|
||||
param_shapes = {k: v for d in param_shapes for k, v in d.items()}
|
||||
|
||||
if debug:
|
||||
for i in range(world_size):
|
||||
print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
|
||||
|
||||
wanted_params = len(param_shapes)
|
||||
wanted_numel = sum(shape.numel() for shape in param_shapes.values())
|
||||
# not asserting if there is a mismatch due to possible padding
|
||||
avail_numel = fp32_flat_groups[0].numel() * world_size
|
||||
print(f"Trainable params: Have {avail_numel} numels to process.")
|
||||
print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
|
||||
|
||||
# params
|
||||
# XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
|
||||
# out-of-core computing solution
|
||||
offset = 0
|
||||
total_numel = 0
|
||||
total_params = 0
|
||||
for name, shape in param_shapes.items():
|
||||
|
||||
unpartitioned_numel = shape.numel()
|
||||
total_numel += unpartitioned_numel
|
||||
total_params += 1
|
||||
|
||||
partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
|
||||
|
||||
if debug:
|
||||
print(
|
||||
f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
|
||||
)
|
||||
|
||||
# XXX: memory usage doubles here
|
||||
state_dict[name] = torch.cat(
|
||||
tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)),
|
||||
0).narrow(0, 0, unpartitioned_numel).view(shape)
|
||||
offset += partitioned_numel
|
||||
|
||||
offset *= world_size
|
||||
|
||||
# Sanity check
|
||||
if offset != avail_numel:
|
||||
raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
|
||||
|
||||
print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
|
||||
|
||||
|
||||
def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states):
|
||||
state_dict = OrderedDict()
|
||||
|
||||
# buffers
|
||||
buffers = zero_model_states[0].buffers
|
||||
state_dict.update(buffers)
|
||||
if debug:
|
||||
print(f"added {len(buffers)} buffers")
|
||||
|
||||
_zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
|
||||
|
||||
_zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
|
||||
|
||||
# recover shared parameters
|
||||
for pair in zero_model_states[0].shared_params:
|
||||
if pair[1] in state_dict:
|
||||
state_dict[pair[0]] = state_dict[pair[1]]
|
||||
|
||||
return state_dict
|
||||
|
||||
|
||||
def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None):
|
||||
"""
|
||||
Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
|
||||
``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
|
||||
via a model hub.
|
||||
|
||||
Args:
|
||||
- ``checkpoint_dir``: path to the desired checkpoint folder
|
||||
- ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
|
||||
|
||||
Returns:
|
||||
- pytorch ``state_dict``
|
||||
|
||||
Note: this approach may not work if your application doesn't have sufficient free CPU memory and
|
||||
you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
|
||||
the checkpoint.
|
||||
|
||||
A typical usage might be ::
|
||||
|
||||
from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
|
||||
# do the training and checkpoint saving
|
||||
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
|
||||
model = model.cpu() # move to cpu
|
||||
model.load_state_dict(state_dict)
|
||||
# submit to model hub or save the model to share with others
|
||||
|
||||
In this example the ``model`` will no longer be usable in the deepspeed context of the same
|
||||
application. i.e. you will need to re-initialize the deepspeed engine, since
|
||||
``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
|
||||
|
||||
If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
|
||||
|
||||
"""
|
||||
if tag is None:
|
||||
latest_path = os.path.join(checkpoint_dir, 'latest')
|
||||
if os.path.isfile(latest_path):
|
||||
with open(latest_path, 'r') as fd:
|
||||
tag = fd.read().strip()
|
||||
else:
|
||||
raise ValueError(f"Unable to find 'latest' file at {latest_path}")
|
||||
|
||||
ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
|
||||
|
||||
if not os.path.isdir(ds_checkpoint_dir):
|
||||
raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
|
||||
|
||||
return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir)
|
||||
|
||||
|
||||
def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None):
|
||||
"""
|
||||
Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
|
||||
loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
|
||||
|
||||
Args:
|
||||
- ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
|
||||
- ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin)
|
||||
- ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
|
||||
"""
|
||||
|
||||
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
|
||||
print(f"Saving fp32 state dict to {output_file}")
|
||||
torch.save(state_dict, output_file)
|
||||
|
||||
|
||||
def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
|
||||
"""
|
||||
1. Put the provided model to cpu
|
||||
2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
|
||||
3. Load it into the provided model
|
||||
|
||||
Args:
|
||||
- ``model``: the model object to update
|
||||
- ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
|
||||
- ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
|
||||
|
||||
Returns:
|
||||
- ``model`: modified model
|
||||
|
||||
Make sure you have plenty of CPU memory available before you call this function. If you don't
|
||||
have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
|
||||
conveniently placed for you in the checkpoint folder.
|
||||
|
||||
A typical usage might be ::
|
||||
|
||||
from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
|
||||
model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
|
||||
# submit to model hub or save the model to share with others
|
||||
|
||||
Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
|
||||
of the same application. i.e. you will need to re-initialize the deepspeed engine, since
|
||||
``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
|
||||
|
||||
"""
|
||||
logger.info(f"Extracting fp32 weights")
|
||||
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
|
||||
|
||||
logger.info(f"Overwriting model with fp32 weights")
|
||||
model = model.cpu()
|
||||
model.load_state_dict(state_dict, strict=False)
|
||||
|
||||
return model
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
|
||||
parser = argparse.ArgumentParser()
|
||||
parser.add_argument("checkpoint_dir",
|
||||
type=str,
|
||||
help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
|
||||
parser.add_argument(
|
||||
"output_file",
|
||||
type=str,
|
||||
help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)")
|
||||
parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
|
||||
args = parser.parse_args()
|
||||
|
||||
debug = args.debug
|
||||
|
||||
convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, args.output_file)
|
||||
Reference in New Issue
Block a user